Implementing SaaS Solution for CRM
Adriana LIMBĂŞAN, Lucia RUSUFaculty of Economics and Business Administration, Babeş-Bolyai University, Cluj-Napoca
[email protected], [email protected]
Greatest innovations in virtualization and distributed computing have accelerated interest in cloud computing (IaaS, PaaS, SaaS, aso). This paper presents the SaaS prototype for Cus-tomer Relationship Management of a real estate company. Starting from several approaches of e-marketing and SaaS features and architectures, we adopted a model for a CRM solution using SaaS Level 2 architecture and distributed database. Based on the system objective, functionality, we developed a modular solution for solve CRM and e-marketing targets in real estate companies.
Keywords: E-Marketing, SaaS Architecture, Modular Development Introduction
Internet marketing is the process of pro-moting an organization by means of the me-dia and the online advertising panelsin order toincrease profit or meet management objec-tives.
E-Marketing requires a comprehensive strat-egy implementing the business and sale model of a company on a web page with its respective functionality and interface. The strategy needs to be adapted to the specific market and centered on an appropriate adver-tising mean (online, media) and using an ap-propriate design [1].
The e-marketing advantages bring to a busi-ness can be pointed at source level: market development, reduces the costs of the cus-tomer and opens new communications me-thods with clients. It developsthe market, as it ensures a larger geographic coverage of the sales and opens markets otherwise inaccessi-ble. It also promotes the label of a company in activity fields in which it is advertising, but also in fields in which it is not actually present. It reduces the costs of the customer service which allows clients to get informa-tion on the products and prices online and to communicate online with the representatives of the customer service in taking the decision of purchasing. It opens new communications methods with clients as e-marketing activities improve the information flow towards the online community [2].
As a new technique in software technology Software as a Service (SaaS) is assimilated with on-demand software service, provided by the vendors to consumers. It enables to hire any software application only when there is a requirement of such a utility. The licensing for using this kind of application may be with a single user or it can be shared with multiple users and offers a simple and economic way to have proper software facili-ties with a minimum of expenses.
The SaaS methodology became more popular in crisis period, because of its high services and less maintenance and costs. The availa-bility on demand makes is highly flexible when the maintenance of the tool is taken in-to account. SaaS saves the complex hardware requirements, once the use of the software is over and enables a centralized control of the business by the service provider. The net-work of action is distributed amongst many users from a single server and offers sharing in the license. SaaS also omits re-installing of the software within an organization because of the central control.
Like End User License Agreement (EULA), sharing of license by SaaS within an organi-zation makes companies cut down the ex-penditure on and offers individual security because there are no copies to distribution, which further extinguishes threats on security of individual system. SaaS is the perfect ser-vice to meet the demands of any software consumer [3].
This paper is structured as follows: after an introduction about e-marketing and SaaS in virtual organizations, Section 2 focused on SaaS features and characteristics. In Section 3 we present CRM prototype as a SaaS solu-tion for real estate companies. We concern on the role of data storage in e-marketing, and then present a dynamic SaaS prototype. We start with architecture model and func-tionalities, then we point Mason features for dynamic reports and personalized query in back-end part of application. Section 4 con-tains conclusions about main benefits of project and the future work. The paper shall include an introduction on the current re-search in the papers field, original solutions, experimental results analysis, conclusions and references.
2 SaaS features and characteristics
More than a revolutionary technology, cloud computing is a relatively new concept, be-cause various types of clouds designed for production-ready enterprise applications and hybrid computing [4].
Cloud computing is “a general term for any-thing that involves delivering hosted services over the Internet”. These services are divided into three categories: Infrastructure-as-a-Service (IaaS), Platform-as-a-Infrastructure-as-a-Service (PaaS) and Software-as-a-Service (SaaS) [5]. A cloud service has three distinct characteristics that differentiate it from traditional hosting: it is sold on demand, (by the minute or the hour), it is elastic, because a user can have as much or as little of a service as they want at any given time, and the service is fully ma-naged by the provider, the consumer needs only a personal computer and Internet access. The goal of cloud computing is to provide easy, scalable access to computing resources and IT services [6]. A public cloud sells ser-vices to anyone on the Internet and a private cloud is a proprietary network or a data cen-ter that supplies hosted services to a limited number of people. Amazon Web Services is the largest public cloud provider. A service provider can use public cloud resources to create their private cloud, and obtain a vir-tual private cloud [5].
Infrastructure-as-a-Service (IaaS) provides virtual server instances with unique IP ad-dresses and blocks of storage on demand. Customers use the provider's application program interface (API) to start, stop, access and configure their virtual servers and sto-rage. Amazon Web Services is the significant example of IaaS [7]. Cloud computing al-lows a company to pay for only as much ca-pacity as is needed, using pay-for-what-you-use model, and bring more online as soon as required.
Platform-as-a-service (PaaS) is defined as a set of software and product development tools hosted on the provider's infrastructure. PaaS providers may use APIs, website portal or gateway and developers create applica-tions on the provider's platform over the In-ternet. Several PaaS examples are Force.com, Salesforce.com and GoogleApps. There are no standards for interoperability or data portability in the cloud and some pro-viders will not allow software created by their customers to be moved off the provid-er's platform.
In the software-as-a-service (SaaS) cloud model, the vendor supplies the hardware in-frastructure, the software product and inte-racts with the user through a front-end portal. Service provider hosts both the application and the data, and the end user can use the service from anywhere. SaaS is a very broad market because services can be anything from Web-based email to inventory control and database processing.
As a part of business modeling concept of cloud computing, SaaS can help in several ways: controlling software licensing ex-penses, controlling rogue software installa-tions, reducing infrastructure expenses. Con-trolling software licensing expenses are re-duced by utilizing a software service provid-er licensing, patches, upgrades, and renewals are tightly monitored. A company pays for what it needs as software elasticity. A soft-ware service provider allows the company to establish an approved applications list and keep it enforced. Also IT department no longer has to support random applications specific to one of users and streamlining
ap-plication support improves efficiencies, ex-pertise and keeps everyone working and con-trol rogue software installations. Infrastruc-ture expenses are reduced based on Web-based application access, which allows com-panies to purchase only the amount of desk-top horsepower needed for the job. Entire business suite of applications is installed on SaaS provider and for many roles, thin desk-tops and virtual operating systems are enough [8].
SaaS architecture is classified by Microsoft from primary maturity level 1 to 4, using several key attributes of the application: con-figurability, scalability and multi-tenant effi-ciency. The four levels are distinguished from each other by this attributes [9]:
• Level 1 – The Ad-Hoc/Custom: Every
customer is equipped with its own modified version of a hosted application and is solely responsible to run their own instances of an application on the servers of the host. The ef-fort required by the SaaS to transfer the non-networked in this level for the development is minimal and the operating cost is even re-duced. The reduction is primarily due to con-solidation of server hardware & administra-tion.
• Level 2 – Configurable: SaaS offers
greater flexibility and in this level, customers are provided with the metadata which is con-figurable by the user and this enables many customers to use separate cases of similar application codes. Various needs of the cus-tomers can easily be offered by the vendor or provided the customer with the detailed con-figuration options. Maintenance and updating of the common code bases are made easy and simple for the users.
• Level 3 – Configurable &
Multi-Tenant-Efficient: In this level, we have lacks scalability but multi-tenancy makes way for the single program instance for serves each customer. Serves resources can be used to maximum.
• Level 4 – Scalable, Configurable, and
Multi-Tenant-Efficient: This is the final ma-turity level, with all of the key attributes. It provides a complete balance of the identical application cases, which are running on the
variable count of servers and offers scalabili-ty. The capacity of the system can be varied according to the demand (by adding or taking away servers) and no changes have to be made in application software architecture [10].
One more key attribute, which may be used in SaaS architectures, is virtualization which replace multi-tenant attribute. The advantage of adding up virtualization over multi-tenant application in the architecture is that the sys-tem’s capacity can be increased with no fur-ther programming. If both attribute are present, SaaS provide greater flexibility for tuning the system for best performance.
3 CRM Prototype – a SaaS approach 3.1 Role of data storage in e-marketing
Companies constantly present on the market record information every time a client gets in touch with one of the departments of the firm. Contact points can be: an acquisition made by a client, a phone call at the clients request, an information enquiry in the online system or a sale card sent by mail. Client re-lated details can be drawn out of several sources. In order to do so, a process of ga-thering, storing and analyzing the data is ne-cessary. The data is used to get information on the clients. It stays as basis for future marketing activities, using the technology of data storage. In order for data base marketing to be used it must recoup the investment. The process of data gathering consists of several elements:
• Elements for extracting and transforming – for processing the data necessary for important applications;
• Tools for sorting data – for detecting and removing the data that is inaccurate, out of date, incomplete, redundant or in in-appropriate format.
• Tools for moving the data – changing the storage place of the data (extraction, transformation and loading)
• Tools of data storage
• Tools for accessing data – for the ex-tracting, viewing, manipulating, processing and presenting the data.
• Tools for delivering data – for commu-nicating, storing and extracting data safely and providing it to the final user [11].
All data that is gathered and processed by means of one of the above mentioned instru-ments is centralized in the contact unit of the firm. Thereafter, the information is organized as data storage.
Table 1. Marketing objective and on-line indicators Marketing
objectives
Indicators
Sale increase S1. Sales – by visitors, by view S2. Sales – done online or by phone S3. Conversion rate (sales/visitors) S4. Purchase trend (hour/day/place)
S5. Entry (from purchase browser, from a link from another site) S6. Exit
S7. Click flow Post-sale
service
Ps1. View of FAQ page Ps2. Downloading files Ps3. Duration of visit Generating
clients’ data
C1. Conversion rate (information/visitors)
C2. Time spent on the site (prospecting and reviewing of the product) C3. Depth of visit
Brand development
B1. Number of visits/visitors B2. Time spent on the site B3. Depth of visit
The employees can register the data, search it according to different criteria and analyses it in order to draw useful conclusions regarding the needs and reactions of an individual client. The information offers a complete im-age on the relation with the client and can be subsequently used, for example, by telemar-keters to answer questions. The marketing department is responsible with determining the extent to which the online objectives were met a proper indicators associated with it (Table 1).
Therefore, we can acknowledge the impor-tance of a database for getting truly signifi-cant results that will assure a good develop-ment of the marketing activity done online. It can also provide success and continuity for the online business [12].
3.2 Architecture and functionality
A relative large number of teams work to-gether to develop the application and, as a re-sult, an UML diagram components was
needed, to help identifying and modeling the software components and their interfaces. Once the interfaces are defined and accepted by the teams, it becomes easier to manage the work of smaller teams.
The back-end application represents a solu-tion developed by the studied company based on SaaS principles and it is used by users having the status of administrators, in order to assure a more efficient management of the site. Our proposal is based on SaaS Level 2 of maturity, offering for real estate compa-nies or brokers as suitable customer confi-gurable metadata, allowing user to fit the in-formation to their needs. Also our model enables many customers to use separate cases of similar application codes for every e-marketing campaign. Various needs of the customers can easily be solved by developing Availability and Performance Modules. In addition we provide the customer with the detailed configuration options (Figure 1).
Fig. 1. Components diagram
This is divided in specialized modules for certain groups of users, on a certain domain of interest. For example, a module can offer a range of tools to the programming team that allows adding, deleting or updating compo-nents in the front-end section, as well as monitoring current tasks or the status of those that are in production. It also provides tools for statistics or customer service and an inter-face that facilitates the best practice of online marketing. Its functionality is based on the collaboration and the dependencies between the interface and the user, on one hand and with the databases, through the communica-tion module, on the other hand, in order to obtain optimized results.
For the back-end section, the actors are con-sidered to be the company’s marketing de-partment members. An actor has the status of administrator, with unlimited permission rights for reading, updating or writing files. He can access the application based on
user-name and password. The main objective is to streamline the business process: to attract a high number of visitors on the site and to convert them into customers, by creating ac-counts and subscriptions.
In order to obtain the desired results, a sec-tion was created for the online marketing ac-tivity. It has the following options:
• Identifying new clients for later target-ing;
• Access to statistics about the clients’ da-tabase;
• Extracting client lists based on specific criteria, for customizing the e-mail cam-paigns;
• Access to the records with the tasks completed by the marketing team and published by the press;
The corresponding scripts select, mostly, data about users such as personal information (first name, last name, and location), email address, profession or domain of interest, as
well as information concerning the account’s type and the subscriptions they own. Also, these scripts use tables that store data about the properties listed on the site (building type, address, sale price, previous sales).
3.3 Using Mason for dynamic reports
Mason is a mechanism for embedding Perl code into HTML, using a particular set of choices when selecting the method to inte-grate the code. This is one of the reasons that make Mason very practical and straightfor-ward. Mason was designed to simplify the process of creating, managing and maintain-ing complex sites or other groups of dynami-cally generated documents. It fully coope-rates with Perl code, improving its popularity and success. It encourages the programmers to see the site in structural terms and not as a collection of scripts and procedural modules. The development of dynamic and complex sites represents the most common application in which Mason is applied. It can be used in almost any situation that implies mandatory control of the content, such as generating mail-merged from letters, creating files with custom configuration and even building dy-namic GIF images, based on varying input parameters.
Mason provides multiple improvements in terms of templating systems written in Perl, using some features aimed at helping man-age, in a more effective way, the building process of a site and its maintenance. The main features are:
• Components – modular design elements. A component is the basic unit of Mason that can accept input parameters and generate output results. Mason allows any component to call another compo-nent at any point during its execution in the same manner that Perl does (a sub-routine can call another subsub-routine). A component may represent a single web page (for example, a navigation bar) or even a shared function that generates no output on its own.
• Object inheritance – it is considered to be one of the Mason’s most important features and allows a component to inhe-rit the behavior of other components, in
the same matter that classes and object do in object-oriented hierarchy. In this way, every component can inherit from another component called auto-handler. The auto-handler implements the com-mon behavior for all the components, for example, the content of header and foo-ter sections. Individual components im-plement specific behavior, such as the body text of every single page.
• Intelligent caching mechanism – when working with dynamic sites, there are situations when certain portions can take longer to generate or they are semi-dynamic (its content is periodical up-dated, but remains static for long periods of time between the changes). In these cases, Mason provides a very sophisti-cated and complex caching mechanism that can be used to control how often the output of a component is rebuilt. The expiration decision is based on time, on certain key parameters (username or content) or an explicit agent can decide when a specific data expires [13].
3.4 E-marketing tools for marketing anal-ysis and customer targeting
When adopting email campaigns as a pro-moting strategy for a business, it is essential that the targeted population is represented only by interested users. For a more accurate selection, the marketer has the possibility to use a wide range of tools that provide him data about both new and old users from the company’s database. A first option allows him to generate a list with the new users on site. The search takes into consideration a specific date or a custom time range [14]. In order to obtain the proper results and, after that, to export them into an excel file, the ap-plication uses a query based on certain key criteria: a field that identifies if a user had a previous account and a subscription. Then, the results are exported into an Excel file and filtered to obtain a more specific targeted segment [15].
Statistics and other information about users offer the marketer the possibility to visualize the results as a list or a chart, grouped by specified fields: profession, primary region,
and secondary region, both primary and sec-ondary regions. The graph is generated using a component called by the page that receives as input parameters the axis coordinates. The custom search is useful if you want to gener-ate rapidly a list with clients that make the subject of your campaign. In this case, the marketer can track issues related to the re-gion he targets, the users’ profession or the type of his account / subscription [16].
For example, if the campaign’s purpose is to generate traffic on the site from users inter-ested in residential properties listed on the X
city’s market, you have to choose a location based search (Primary Region) and to select the corresponding profession (homebuyer). The script runs a query on the involved tables, with the conditions mentioned before and returns a list with the results that can be exported as a .CSV file. The results are gen-erated based on the logical OR operator be-tween the two sections and the key words are limited as it follows: the comma is the equiv-alent for “OR” and the space for “AND”. We used extended database from abroad because Romanian databases is not ready yet (Fig. 2).
Fig. 2. Custom search
A new section (Press) was created and its aim is to assure a more effective method for managing and monitoring the marketing de-partment tasks that involve working with da-tabases. This section keeps track of the activ-ity developed by each employee, both for in-ternal use and for press publications (online or written). Each marketer enters his task, specifying the following elements: the publi-cation type (online, written, and other), a short description, the request date, the pre-pared date, the authors name and must upl-oad the result file. The page is divided into two sections: the press request page that in-cludes analysis made based on the requests received from different publications and the
press mentions page that records the articles in which the company was cited as data source [17].
The marketing department has to deal every day with a large number of requests concern-ing the real estate market. An automation of the gathering process reduces the time needed to complete a task and, also, the re-dundancies.
We developed an accessible tool for every marketer, even if he does not have databases knowledge. This page is based on databases querying principles and involves the tables that record information about the properties listed on the market. After selecting the time range and the region of interest, the
differ-ence between the initial listing price and the current price is calculated. It provides a list with top 30 price changes in both ways: in-crease and dein-crease.
4 Conclusions
The online promoting tends to become the most convenient marketing media for a com-pany that helps it to enlarge its distribution channels for products and services. This is the point where integrating the databases in the process can make the difference between two companies.
The maintenance of a database with informa-tion about the consumers, their preferences and a track of their behavior can bring the following advantages for a company:
• Customizing e-mail campaigns based on specific target audience, which avoids mass promotion of products and services and address each class of customers ac-cording to their needs;
• Instant communication with the client, an online marketing strategy that can turn any online business into a real source of profit;
• Availability in time and customer feed-back that allows him to react, express his opinions or send suggestions to the com-pany;
• Real-time statistics for measuring the success of promotional campaigns,
ob-tained by querying the databases and a variety of other tools at the employees’ disposal;
• The efficiency increase due to the consi-derable reduction of time by selecting a target segment for a given action and ob-taining lists of thousands of customers who meet the requirements and will be the destination of the promotion cam-paigns;
The future development directions for a Business Intelligence module take into con-sideration:
• Extent of the searching criteria based on the product point of interest;
• Developing of the monitoring section for the marketing department;
• Implementation of a system that moni-tors and extracts statistics on income generated from the email campaigns; • Implementation of new tools to achieve
statistics on real-estate market;
• Making a section of HCI (Human Com-puter Interaction) to create customized popups by any team member.
Acknowledgment
This work was supported by Romanian Na-tional Authority for Scientific Research un-der the grant no. PN2 92-100/2007 SICO-MAP.
References
[1] P. Kotler and K. L. Keller, Marketing Management, Teora Publishing House, Bucureşti, 2008.
[2] P. Kotler and G. Armstrong, Principles of marketing, Prentice Hall Europe, 1999. [3] SaaS “New Era of Software”,
www.neweraofsoftware.com/saas.aspx, last access 20.05.2010.
[4] Rss,
<http://www.techtarget.com/180.xml>, last access 9.01.2010.
[5] Cloud Computing, http://WhatIs. techtar-get.com/definition/0,,sid9_gci1381649,00 .html, last access 19.03.2009.
[6] Cloud_Computing –Definition,
http://searchCloudComputing.techtarget.c
om/sDefinition/0,
,sid201_gci1333074,00.html, last access 19.03.2009.
[7] Cloud_Computing, http://searchCloud Computing.techtarget.com/sDefinition/0, ,sid201_gci1356516,00.html, last access 19.03.2009.
[8] Cloud_d, 2009 http://searchExchange. techtarget.com/sDefinition/0, ,sid43_
gci213778,00.html, last access 19.03.2009.
[9] ItTools,
www.ittoolbox.com/wiki/index.php/SaaS , last access 30.03.2010.
[10] SaaS, http://www.whatissaas.net/ saas-software-as-a-service.html, last access 5.06.2010.
[11] Şt. Niţchi, R. Avram Nitchi et all, Elemente de baze de date şiprogramareaplicateîneconomie, Riso-print Publishing House, Cluj-Napoca, 2006.
[12] P. Kotler, Ten deadly marketing sins, Ed. John Wiley and Sons, New Jersey, 2004.
[13] Mason Developer’s Manual,
http://www.masonhq.com/docs/manual/D evel.html, last access 21.04.2010.
[14] SaaS Definition, http://searchCIO-Midmarket.techtarget.com/sDefinition/0, last access 5.06.2010.
[15] SaaS, http://searchNetworking. tech-
tar-get.com/sDefinition/0,,sid7_gci212176,0 0.html, last access 5.06.2010.
[16] SaaS concept, http://www.wikinvest. com/concept/Software_as_a_Service, last access 5.06.2010.
[17] SaaS development, http://www.saas-development.com, last access 5.06.2010.
Adriana LIMBĂŞAN has graduated in 2008 at Babeş-Bolyai University of
Cluj-Napoca, Faculty of Economics and Business Administration, Business Information Systems Department, and from June 2010 is e-Business Master. Nowadays she is marketing analyst at Property Shark company, which pro-vides real estate professionals, investors and homebuyers with comprehensive property information. The competencies covers: extracting and preparing data to meet custom requests for press and blog posts, SEO, checking the data accuracy on our site and from other sources, preparing monthly and quarterly market reports, preparing data for email campaigns.
Lucia RUSU, PhD, Professor at Babes-Bolyai University of Cluj-Napoca,
Business Information Systems Department from 2001. She is graduate in Computer Science and Automatic Control, Technical University of Cluj-Napoca, and Doctor in Economy, Business Information Systems domain from 1997. Her research interests: graphics and web programming, multime-dia and distributed technologies, integrated systems EAI and ERP, workflow management systems, processes modeling and management, mobile applica-tions development. She was member in several research grants in several domains: web appli-cations management, quality management, collaborative systems, workflow management systems.
